ground

I mean... isn't that part of the point of benchmarks?

https://hwbot.org/submission/4468838_terraraptor_wprime___32m_core_2_duo_e8200_13sec_219ms

4720 on 2x16 on a 4 dimmer, impressive! Nice chip too!

2 new AMD 32M records in a day, not something you see every day...

been playing some max memory frequency on x58, Gdie+x58a-oc feels like cheating. http://valid.x86.fr/81dluy 1.5V VTT, 2.2V Vdimm + mostly manually tweaked timings Might hook up the peltiers in a couple days and hope the memory scales further... I'm pretty sure I'm trp limited now.

Any chance I can get rid of the numbers? So its just Ground?

I still wish platform records were worth something as well

Impressive!

stability is pretty much all gone over ~1650, and dual channel is impossible. This is only possible due to the x58a-oc allowing to set lower uncore clock if you use the innermost channel. It is possible to run 16x mem + 17x uncore this way. This CPU tops out around 1400 32M on R3BE with typical uncore ratios and is the strongest IMC I’ve found on this platform by a fair amount. So far I’m only running some timings manual, I recon ~30-50 MHz more might be doable without increasing volts by a lot, since with full manual timings since gdie appears quite sensitive to that, at least on z87 this stick did. I also have 2 more sticks to try as well as a couple more tweaks I’ll be keeping to myself for now.

Gdie just seems like plain magic for memory validation :D Guess I need to try again...

@T.Rex We have so much to bench when I get to denmark again But yeah I was planning on bringing this chip for sure, 1800+ is the goal The mem can do it. Thanks whiteshark! (for context - this memory frequency score is done on x58 https://hwbot.org/submission/4611155_ground1556_memory_frequency_ddr3_sdram_1645_mhz)

maybe its a monitor with integrated PS/2 ports and thus its just a weird VGA cable? The card looks like some generic PCI GPU to me...

3:5 is a broken divider, 2:3 real, so 645*3/2 = 967 real which is still quite nice of course. Lots of 775 scores with misleading memory frequency, but some boards just run best at this divider so can't do much about it either way. This also means that every couple months someone has to sift through the DDR2 frequency rankings to check if any subs were made with the broken divider. Great chip you have there!

@flanker I was testing both BIOS versions on P5E Deluxe a while back and Maximus Formula topped out around ~520 with Wolfdale (E8400) while Rampage Formula BIOS did ~610 with everything else exactly the same (or as similar as possible).

https://www.ebay.de/itm/DDR2-DDR3-Desktop-RAM-Testkartenadapter-Speichertester-Analyzer-W-LED-Anzeige/163700848617? These are quite useful for debugging memory slots.

the IC is only noted in very few DDR3 kits. Beyond that, 8GB sticks aren't gonna be Rev D anyways since its a 2Gbit ic, so largest sticks containing it (not counting registered memory) are 4GB sticks, like 2666c10 dominators, 2600c10 gskill tridents, 2400c9 tridents and similar.

yeah, who dares covering part of the window? thats just wrong...

Not a fun chip to bin for sure, congrats!

I think I've seen that issue before with DFI 775 boards

Memory on Ln2 as well?

https://xkcd.com/1172/

I'm intrigued and slightly terrified.

There the 7700K/7740X reign ends...

I've never tested actual current draw, but I recon a 2-3 phase setup could be useful for smaller GPUs, maybe x48 chipsets and x58 VTT rail (never actually checked current draw of that rail, need to test at some point). There are probably more usecases that I can't think off right away.

534 fsb 32M with a 45nm quad on x48, wow. Good run!